David King

MLX Dabbler  • Cardano HODLer ₳ • Vision Pro ᯅ • Junkie • π

[David King @theuttermost]

@inco_ai @ornith_ How do we run this? I am getting the following error in Splash.

error: no supported model has this architecture (text config hidden_size: GGUF 3072, Qwen3.6-35B-A3B 2048); supported: Qwen3.8-27B, Qwen3.6-35B-A3B, Qwen3.8-Flash-Next

[David King @theuttermost]

@jundotkim Thank you so much for including Splash as an engine! We can now run GGUFs! Let's go!

[David King @theuttermost]

@Teknium @HermesAgentTips 👀... Let's gooo!!

[David King @theuttermost]

@udiWertheimer @markgurman I NEED THIS!!!!!!!

[David King @theuttermost]

@ivanfioravanti @twid This is what I am doing! Waiting to get 2xM5 Ultra's

[David King @theuttermost]

@ivanfioravanti Exact same here... I should have gotten the 256GB

[David King @theuttermost]

@dhh @0xSero Wow!! I have the utmost confidence in @0xSero! I might just have to ditch Mac OS... Can't believe I am even saying that.

[David King @theuttermost]

@Teknium Let's gooooooo!!!! I love you guys so much!

[David King @theuttermost]

Can't say how happy I am to see this!! Big congrats to @jundotkim! The future is bright!

[Jun Kim @jundotkim]

I'm happy to announce that I've joined Hugging Face.
What started as a personal project back in February is now something I get to work on full time.

Local AI has grown explosively this year. I've said this since the early oMLX releases: I want my friend who bought a MacBook yesterday to be able to run AI on it today. I believe MLX has that potential. Apple Silicon is the easiest entry point for a regular person to get started with AI, with no complicated hardware to assemble. And the open source community, including Hugging Face, has been growing that potential. That support has already made a huge difference.

Hugging Face is the best place for me to support oMLX and the MLX community with everything I have. For anyone getting into local AI, the first step usually starts at Hugging Face. Mine did too. I'm proud that I get to work at that entry point, where new ideas can be tried out.

oMLX stays exactly where it is, under the same Apache 2.0 license in the same repository, and I'll keep leading the project, same as before. What changes is that I can now spend far more time on it, move faster, and build something sustainable for the long term together with all the contributors who have put so much into it.

I also want to do more for MLX as a whole. oMLX is built on top of transformers, mlx-lm, mlx-vlm and the rest of the ecosystem, and I'm grateful to the people behind them. Rather than keeping everything inside oMLX, over time I want to push work upstream where it makes sense. And where the community needs something that doesn't exist yet, oMLX is a good place to try it first.

Thank you to the 264 contributors who have built oMLX with me, and to everyone who filed issues with detailed logs and reproductions so we could fix things. oMLX would not be what it is without you.

oMLX continues in the same place, in the same way. Just much faster.

HF's announcement: https://t.co/LjOGKGWw64

[David King @theuttermost]

@0xRishi Bruh.... I legit couldn't tell the difference between this and the real thing... Where's the link? I wanna play.

[David King @theuttermost]

@ivanfioravanti @MiniMax_AI @RenLeanna @Ronny_MiniMax What a life!! Love following all your travels!!

[David King @theuttermost]

@ivanfioravanti Agreed. Over the weekend, I was having it OCR images and convert them to markdown, while the speed was awesome the accuracy of the translation was much lower than using @jundotkim oMLX Qwen3.8 oQ4 variant with lighting MTP.

Speed was nice but 4-bit accuracy matters alot.

[David King @theuttermost]

@Prince_Canuma @PrismML @jundotkim @zhijianliu Ohhh nice! Thank you! I typically use via oMLX so maybe I have to wait for an update.

[David King @theuttermost]

@superalesha Thanks for posting this! I def fell for the hype. Will pass on this model.

[David King @theuttermost]

@zhijianliu @BinaryTimeLord Thanks so much for all your work on this, downloading and trying it out now! Super excited.

[David King @theuttermost]

@BinaryTimeLord @zhijianliu Seems to be 4bit quant

[David King @theuttermost]

@ivanfioravanti Super excited to try this, oMLX has been my daily driver for local models, this looks promising. I only really use Qwen 3.8 anyway.

[Inco AI @inco_ai]

Qwen3.8-27B at 144 tok/s on an M5 Max MacBook Pro ⚡

Meet Inco Splash: our open-source inference engine, built around the model and around Apple silicon.

Up to 3× the decode speed of Ollama, 2× oMLX, and almost 4× when an agent fans out into sub-agents.

[David King @theuttermost]

@inco_ai ARE YOU SERIOUS?!?!? I am a daily user of oMLX and love it. Excited to try Inco AI!